Isolation and Characterization of 26 Microsatellite Loci for the Shortfin Silverside Fish Chirostoma humboldtianum Valenciennes1835 (Atherinopsidae: Menidiinae) Derived from Next Generation Sequencing and Their Cross-Amplification in Central Mexican Atherinopsids ()
1. Introduction
Up until the first half of the 20th century, atherinopsids were the main fishing product of the inland waters of Mexico and counted among the top scale fish species. The representative endemic genera of Atherinopsidae in Central Mexico are Chirostoma and Poblana, locally recognized as “charales” and “peces blancos” (silversides), C. humboldtianum included. The species is of great economic importance, because it has been used for decades as food by humans inhabiting the shores of Pátzcuaro and Chapala lakes and reservoirs in the States of Michoacán and Mexico. Although it is true that atherinopsids are of great interest as a substantial screed in epicontinental fisheries capture, it is also true that the development and growth of artisanal fisheries in these regions lack of scientific basis and are rather based on the needs to supply high quality animal protein in short term. The species is composed by two morphotypes, clearly segregated into western and eastern populations and characterized by having differences in size, and in the mean values for median lateral and predorsal scales [1] [2] ; andrecognized as the hypothetical ancestral form of the “humboldtianum group” [1] [3] . During the last decades the populations of the species have been reduced or extirpated due to the introduction of non-native fish to the western basins [1] [2] [4] . Similarly, habitat loss, pollution and overfishing have played important roles in the decline and extinction of local populations of related species (C. riojai, C. promelas, Poblana alchichica) listed in the Official Mexican Standard NOM-059-Semarnat in 2010 and in the IUCN Red List.
A recent study on the population genetics of the species and based on sequences of the mitochondrial control region, described the species as highly differentiated and with intermediate levels of diversity [5] ; and although the authors proposed the conservations of all the populations studied, they also pointed out the possible evolutionary stochastic effects and the limitation of defining conservation units only on the bases of mtDNA. Therefore, eight poymorphic microsatellite loci were developed [6] based on an enrichment method [7] in order to further assess and confirm the diversity and structure of the species. Furthermore and with the aim of developing a larger set of genetic codominant markers available to study the silversides of México, a recently diverged group of fishes [8] [9] , here we characterized 26 additional polymorphic loci for the species and using Next Generation Sequencing and succesfully cross-amplified them in closely related species with the aim of their use in assessing further population genetics and phylogeographic issues that can aid in the conservation and management of atherinopsids.
2. Materials and Methods
2.1. Microsatellite Isolation
Genomic DNA was extracted from muscle tissue of one individual of C. humboldtinuam using a salt extraction protocol [10] and sent to the Georgia Genomics Facility (University of Georgia, Athens, USA) for sequencing. The specimen was deposited in the Colección Nacional de Peces Dulceacuícolas, Escuela Nacional de Ciencias Biológicas-IPN (ENCB-IPN-P6423). DNA was sheared using a Covaris S2, and Illumina adapters were ligated on using methods derived from [11] but using adapters equivalent to Illumina TruSeq with 10 nt indexes [12] . Libraries were pooled and run on the Illumina HiSeq 2000 (paired-end 100 reads). Resulting fastq files were demultiplexed, and reads were filtered and run through the PALfinder pipeline for microsatellite identification and primer design [13] . Over 10 million reads at 100 bp were obtained, covering approximately 1.312 bases and resolving 3× genome coverage. Detected reads with microsatellites were 108, 228, of which 0.16% presented primers. Further primer sorting was as follow: 1) all loci with primers that were found more than once were deleted, particularly most of the really long (>500 bp) and short dinucleotide repeats (<24 bp); 2) removal of the longest repeats. Suitable primer design was possible in 1153 pairs with repeats of mid-length. Fifty eight pairs of primers were selected on the bases of the following parameters: a) length (18 - 30 bp), b) G-C content (40% - 60%), c) melting temperature (55˚C - 65˚C), d) 3’-stability, e) avoid hairpin or dimer formation and f) no selfpriming, and tested for polymorphism.
2.2. Sampling
Mexican government kindly issued permit number DGOPA.07343.310810.4128 to conduct this research. Organisms from Villa Victoria were obtained from commercial artisanal catches. C. riojai and P. alchichica specimens were kindly donated by Gerardo Figueroa Lucero and Héctor Espinoza, respectively. Gill lamella tissue samples from C. promelas, C. sphyraena, C. grandocule and C. jordani were obtained from the Tissue collection of the Laboratorio de Genética y Biología Molecular, Planta Experimental de Producción Acuícola, Universidad Autónoma Metropolitana Iztapalapa.
2.3. DNA Extraction, PCR Amplification and Analysis
Total DNA was isolated as previously described from 30 C. humboldtianum from Villa Victoria Dam, State of Mexico (19˚26'N, 100˚00'W) and 22 specimens of Chirostoma and Poblanaalchichica (Table 1) to test crossamplification.
Table 1. Cross-amplification of 26 microsatellite loci of Chirostoma humboldtianum in five species of Chirostoma and one Poblana species. Sample size in parenthesis.
For C. promelas ++ = 2 genotypes recorded; + = 1 genotype; for C. jordani, C. sphyraena, C. riojai, C. grandoculeand P. alchichica ++ = 3 - 4 genotype products recorded; + = 1 - 2 genotypes; - = no amplification success when n = 4 or n = 2.
PCR were performed in 10 μl reaction, containing 40 ng DNA, 1× PCR buffer (20 mM Tris-HCl, pH 8.4, 50 mM KCl), 0.2 mM dNTPs, 0.2 μM each primer, 2.5 mM MgCl2, and 0.25 - 0.35 U Taq DNA polymerase (Invitrogen, Carlsbad, CA, USA). Cycling conditions included: 95˚C 5 min, 35 cycles of 30 s at 94˚C, 30 s at the locus-specific annealing temperature (Tm) (Table 2), 40 s at 72˚C, and a final 72˚C extension for 15 min and using an iCycler (BIORAD Laboratories, Hercules, CA, USA). PCR products were visualized by silver staining in 6% polyacrylamide gels (7.5 M urea). Allele sizes were determined using 10 bp ladder (Invitrogen). Cross-species PCR amplifications required a reduction on the stringency of reactions on some loci, carried out lowering 1 or 2 grades the Tm.
Table 2. Primer sequences and characteristics of 26 microsatellite amplified in 30 individuals of Chirostoma humboldtianum (Ch) from Villa Victoria dam.
*Significant deviation from HWE (p ˂ 0.05); AN = Accession number; ǂForward primers were 5’ labeled with the M13 sequence (5’TGTAAAACGA CGGCCAGT), for further fluorescence detection.
Potential presence of null alleles was assessed in Free NA [14] . Alleles per locus (NA) and observed (HO) and expected (HE) heterozygosities were calculated using deviations from Hardy-Weinberg equilibrium, linkage disequilibrium was estimated and tested using GENEPOP 4.0.10 [15] under default parameters. Multiple hypothesis tests p values were adjusted by the False Discovery Rate (FDR) method [16] . In order to gain information on the possible utility of these loci in the closely related species assayed, the percentage of polymorphism per species (%) was obtained by accounting only those loci that resolved more than one allele per locus and wihtin each species.
3. Results and Discussion
Among 58 primers tested, 26 microsatellites screened successfully, 25 resolved as polymorphic and Chum53 was monomorphic in specimens from Villa Victoria. Mean number of alleles in the polymorphic loci was 14.3, SD = 9.38, a mean number 2× higher than the resolved earlier for the species (see [6] ; 6.4, SD = 2.5), but similar to those characterized closely related species [17] [18] ; and HO and HE 0.47, SD = 0.32 and 0.58, SD = 0.32, respectively. These later values are similar to those previously reported (see [6] ; 0.61, SD = 0.19 and 0.63, SD = 0.12, for HO and HE, respectively). Linkage disequilibrium was detected for a pair of loci within specimens from the population (Chum15 and Chum39, p = 0.00). Null allele frequency estimates were negligible for half of the loci, moderate (>0.05 and <0.20) for 12 loci (Chum11, Chum14, Chum18, Chum30, Chum33, Chum39, Chum48, Chum59, Chum63, Chum68, Chum69 and Chum70) and high at one locus (Chum27 = 0.357). Significant deviation from HWE was detected in six loci (p ˂ 0.05, Table 2). This deviation might result from the heterozygote deficiency found in some loci (Chum11, 18, 27, 30 & 39), which in turn might be caused by endogamy as has been earlier suggested for the species [5] and in other species of the genus [19] [20] .
Cross-PCR amplifications of 22 individuals of Chirostoma and Poblana resulted in 16 polymorphic loci and the remaining 10 loci resolved genotypes in most of the species (see Table 1). For instance Chum 07, 18, 33, 39, 53, 57, 60, 62, 63 and 69 resolved genotypes in 3 or 4 individuals (++) of each of the six screened species, excepted for C. promelas, for which the number of specimens analyzed was two; Chum 12, 27, 44, 48, 56 and 65, rendered genotypes (++) in five of the species and genotypes (+) in one species per locus. Chum 11, 58 and 68, resolved genotypes (++) in four species and (+) in two species; Chum 14, 30, 55 and 59, also resolved genotypes (++) in four species and (+) in another species, however these loci did not amplified neither in C. promelas or C. sphyraena; the remaining loci amplified genotypes (++ or +) in one to three species and did not amplified in other two or three species. Although eight microsatellite loci have been recently developed for the species (see [6] ), these 26 new microsatellite loci are the first to be tested in cross-amplification between the Mesa Central Atherinopsids and will be helpful tools for genetic population and phylogeographic studies of C. humboldtianum, as well as in paternal assignments for culture purposes, studies all that can aid in developing a conservation plan for this important species, which although is among the group known as “pecesblancos” the species with the widest distribution (see [19] ), some of its populations have been reduced or extirpated due to habitat loss, pollution and overfishing (see [19] ). These markers will also aid in investigating the integrity of stocks/populations for management and conservation of related endangered species.
Acknowledgements
Funding for this research was provided by Consejo Nacional de Ciencia y Tecnología-CB-2009-01-130220 and UAM.147.07.03/147.09.01 granted to IDLABS. This work was completed by Irene D. L. A. Barriga-Sosa during her sabbatical leave in CCA-UAA.
NOTES
*Corresponding author.